Predictions and causal estimations are not supported by the same associative structure

Area of Science:

• Cognitive Psychology
• Learning Science
• Causal Inference

Background:

• Judgments about cue-outcome relationships are influenced by question type.
• Causal link strength relies on cue-outcome contingency, while outcome prediction depends on conditional probability.

Purpose of the Study:

• To investigate the underlying processes influencing judgments about causal and predictive learning.
• To examine the interaction between question type and trial order in causal and predictive judgments.

Main Methods:

• An experiment was designed to test how different question types and trial orders affect causal and predictive judgments.
• Participants' responses were analyzed to identify patterns related to associative and higher-order reasoning.

Main Results:

• An interaction between question type and trial order was observed, challenging existing models.
• This interaction suggests that standard models of associative or higher-order reasoning alone are insufficient.

Conclusions:

• The findings support a dual-process model of learning, integrating both associative and higher-order reasoning.
• Understanding these combined processes is crucial for a comprehensive theory of causal and predictive learning.